ClawHub
Back to skill

Security audit

SEO Keyword Research Tool

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed YouTube transcript helper with no evidence of credential access, persistence, or harmful behavior, though its marketplace name and documentation are somewhat confusing.

Install this only if you want a YouTube transcript summarization helper, not an SEO keyword tool. Run the Python dependency from a trusted package source, consider pinning the dependency version, and ensure the agent treats video captions as content to summarize rather than commands to follow.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises and instructs use of Python code that performs network access to YouTube and reads local script files, but it declares no permissions. This creates a transparency and policy-enforcement gap: users and hosts may approve the skill expecting no privileged actions, while it can still retrieve external content and operate on local resources.

Tp4

High
Category
MCP Tool Poisoning
Confidence
82% confidence
Finding
The skill description promises summarization, configurable summary lengths, and multilingual behavior, but the documented implementation only fetches subtitles and lists available tracks. This mismatch is dangerous because users and orchestrators may rely on claimed behavior and safety boundaries that are not actually implemented, which can lead to unsafe delegation, unexpected data handling, or misuse of a partially implemented skill.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.